2021 14" Mac w/M1 Pro vs. 2019 16" Mac w/i9 - Android Studio benchmarks

[sounds]

Summary

Benchmarked several real world Android Studio tasks

1. 2019 16" Mac w/2.4ghz Intel i9 (8 core)–32gb RAM
2. 2021 14" Mac w/M1 Pro (10 core)— 32gb RAM
3. Desktop (Pop_OS!) — 4.2ghz AMD 2950x (16 core) — 64gb RAM

Android Studio Invalidate Caches and Restart

1. 2021 Mac w/M1 Pro: 97 seconds
2. 2019 Mac w/i9: 177 seconds (82% slower)
3. 2950x: 195 seconds (101% slower)

Consecutive Builds on Battery - shows battery life and thermal throttling:

 

Quotes

Quote

Battery Life

To test battery life, I used Gradle Profiler to do a clean build of our app on repeat with the screen at 1 notch of brightness and nothing else running. This is far beyond normal usage because it keeps nearly all cores running at capacity continuously.

This stress test took my 16" i9 MacBook from 100% to 0% in 58 minutes. An impressively bad result. In that time, it compiled the app 19 times.

The 14" M1 Pro lasted 2 hours 10 minutes. The 16" should offer roughly 30% more battery life which would bring this up closer to 3 hours.

2–3 hours might sound terrible relative to what you might expect from Apple Silicon, however, keep in mind that this test had nearly all 10 cores running continuously.

 

My thoughts

The variation in the build times for the battery life / thermal throttling test looks like pretty strong evidence that it's possible to get the M1 Pro into thermal throttling, if you load it up long enough.

 

Sources

https://gpeal.medium.com/the-m1-pro-for-android-engineers-a144093aa1ec

[SupaKomputa]

im not surprised, that thing got a great single core score. the threadripper pretty much outdated.

[PocketNerd]

12 hours ago, SupaKomputa said:

im not surprised, that thing got a great single core score. the threadripper pretty much outdated.

Yeah, a pretty unfair test on the AMD side of things.

[tikker]

13 hours ago, sounds said:

 

 

Journalists that make graphs should really take a hint or two from the scientific world. In its current form all this graph seems to imply is that the i9 is "better", but can do less builds. The reader doesn't even know what the vertical axis tells you. Label all your axes and put proper captions in there people.

[leadeater]

12 hours ago, SupaKomputa said:

im not surprised, that thing got a great single core score. the threadripper pretty much outdated.

Zen 3 CCX and Cache changes also makes it WAY faster than all Zen archs before, it's one of if not the largest performance workload gains from Zen 2 to Zen 3. Talking about code compiling.

[SansVarnic]

-= Moved to Laptops and Pre-Built Systems =-

Not tech news, this more of a review topic of existing harware.

[Roswell]

20 hours ago, sounds said:

This stress test took my 16" i9 MacBook from 100% to 0% in 58 minutes.

I don't know how the author of the article managed that... something is really wrong with their machine or battery. 16 inch 2019/2020 models have a 100Wh battery and the parts aren't capable of coming close to 100W as far as I know, so something is afoot.